Temporal sensorfusion for the classification of bioacoustic time series [Elektronische Ressource] / Christian R. Dietrich

AbteilungNeuroinformatikProf. Dr. Gunther¨ PalmTemporalSensorfusionfortheClassificationofBioacousticTimeSeriesDissertationzurErlangungdesDoktorgradesDr.rer.nat.derFakultat¨ fur¨ InformatikderUniversitat¨ UlmvonChristianR.DietrichausHirschegg2003iiAmtierenderDekan: Prof.Dr. FriedrichW.vonHenkeErsterGutachter Prof.Dr. Gunther¨ PalmZweiter PDDr. AlfredStreyTagderPromotion 25.06.04AbstractClassifyingspeciesbytheirsoundsisafundamentalchallengeinthestudyof animal vocalisations. Most of existing studies are based on manualinspection and labelling of acoustic features, e.g. amplitude signals andsoundspectra,whichreliesontheagreementbetweenhumanexperts. Butduring the last ten years, systems for the automated classification of ani malvocalisationshavebeendeveloped.InthisthesisasystemfortheclassificationofOrthopteraspeciesbytheirsounds is described in great detail and the behaviour of this approach isdemonstratedonalargedatasetcontainingsoundsof53differentspecies.Thesystemconsistsofmultipleclassifiers,sinceinpreviousstudiesithasbeen shown, that for many applications the classification performance ofa single classifier system can be improved by combining the decisions ofmultipleclassifiers.To determinefeatures forthe individualclassifiers thesefeatures havebeen selected manually and automatically. For the manuale selec tion, pattern recognition and bioacoustics are considered as two coher-entinterdisciplinaryresearchfields.